Improving patient-reported outcomes in cancer care: non-response and the role of proxies

Roydhouse, Jessica

Full Metadata

Overview

Year:: 2017
Contributor:: Roydhouse, Jessica (creator); Wilson, Ira (Advisor); Gutman, Roee (Reader); Mor, Vincent (Reader); Keating, Nancy (Reader); Brown University. Department of Health Services, Policy and Practice (sponsor)
Genre:: theses
Subject:: Cancer; Caregivers; Patient-Reported; Care Quality; Quality of life
Extent:: viii, 106 p.
DOI: https://doi.org/10.7301/Z0J38R06

Files

Description

Abstract:: Patient-centered care is part of care quality. If patients are not able to respond, one commonly used approach is to ask a family member (‘proxy’) to answer on the patient’s behalf. Proxies are asked to substitute for otherwise missing patients in many national and international surveys used for public reporting and/or payment, and inform policy. In many cases, proxy use is infrequent, resulting in little practical impact. In cancer, however, a higher proportion of proxies can be expected. The overall goal of this dissertation is to evaluate methods for using and interpreting patient-reported outcomes when patients are unable to complete questionnaires in cancer. The dissertation focused on care experience, quality, and health-related quality of life. Three questions were addressed: 1) if the inclusion of proxy reports affects the outcomes of interest, 2) if proxy-specific covariates affect how the proxies reported on these outcomes, and 3) if different methods for analyzing proxy responses produce different results for the outcomes of interest. For all three questions, the Cancer Care Outcomes and Research Surveillance Consortium was used for analysis. I also used two datasets for simulation analysis as part of the third question: the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment, which contained health-related quality of life outcomes, and Cancer Assessments and Reports of End of Life Treatment, which contained experience/quality outcomes. Our findings suggest that the use of proxy responses produces small differences for experience and quality outcomes, but large, clinically significant differences for quality of life outcomes. Furthermore, proxy characteristics, especially the proxy’s involvement in patient care, are important predictors of proxy reports of patient outcomes. Furthermore, inverse probability weighting and multiple imputation had limited usefulness for accounting for proxy responses in this dataset. While rank permutation appeared to be a promising approach, the simulation findings suggested that rank permutation performs best if patient data are missing due to observed covariates. These findings suggest that 1) proxy reports do have an impact, particularly for quality of life, 2) additional information should be collected about proxies, and 3) rank permutation should be considered when analyzing proxy-reported data.
Notes:: Thesis (Ph. D.)--Brown University, 2017

Content

Access Conditions

Rights: In Copyright
Restrictions on Use: Collection is open for research.

Citation

Roydhouse, Jessica, "Improving patient-reported outcomes in cancer care: non-response and the role of proxies" (2017). Health Services, Policy & Practice Theses and Dissertations. Brown Digital Repository. Brown University Library. https://doi.org/10.7301/Z0J38R06

Relations

Collection:

Health Services, Policy & Practice Theses and Dissertations

Theses and Dissertations for the Health Services, Policy & Practice department.
...